BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Machine Learning Reading Group @ CUED
SUMMARY:Completely Random Measures in Bayesian Nonparametr
ics - Dr Daniel Roy (University of Cambridge)\, Cr
eighton Heaukulani
DTSTART;TZID=Europe/London:20121018T143000
DTEND;TZID=Europe/London:20121018T160000
UID:TALK41135AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/41135
DESCRIPTION:In Bayesian nonparametric modelling\, we allow a m
odel to learn an unbounded number of parameters by
replacing classical finite-dimensional \nstatisti
cal distributions with infinite-dimensional stocha
stic processes. Completely random measures are a
special class of stochastic processes\, \nwhich ar
e intuitive\, highly interpretable\, and useful fo
r statistical applications. While this theory is
mathematically deep\, in this tutorial we will ins
tead \nfocus on practicalities such as how to samp
le such objects\, providing pointers to more theor
etical material when necessary. This tutorial wil
l include an \nintroduction to random measures and
their practical uses\, Poisson processes and how
to sample them\, complete randomness\, and how to
characterise \na completely random measure. Final
ly\, we introduce some important completely random
measures\, such as the Bernoulli process\, the Be
ta process\, \nand the Gamma process\, and we will
show how they are related to some popular objects
used in machine learning\, such as the Dirichlet
process\, \nChinese restaurant process and the Ind
ian buffet process.
LOCATION:Engineering Department\, CBL Room 438
CONTACT:Konstantina Palla
END:VEVENT
END:VCALENDAR